Published October 1, 1994
Induction of Polarized Cell-Cell Association and Retardation
of Growth by Activation of the E-Cadherin-Catenin
Adhesion System kn a Dispersed Carcinoma Line
M i t s u k o Watabe, A k i r a Nagafuehi,* Saehiko Tsuldta,* a n d M a s a t o s h i Takeiehi
Department of Biophysics, Faculty of Science, Kyoto University, Kitashirakawa, Sakyo-ku, Kyoto 606-11, Japan; and * Department
of Information Physiology, National Institute for Physiological Sciences, Myodaiji, Okazaki 444, Japan
Abstract. PC9 lung carcinoma cells cannot tightly as-
ELLS of simple epithelia have apical-basal polarity,
which condition is essential for their structure and
function (Rodriguez-Boulan and Nelson, 1989).
Most of simple epithelial cells are linked together by a junctional complex comprised of fight junction, adherens junction (zonula adherens or intermediate junction) and desmosome, which are arranged in this order from the apical
portion of the cell-ceil contact (Farquhar and Palade, 1963).
The basal plasma membrane of the cells attaches to the basement membrane, while the apical-free surface develops
microvilli for responding to the extracellular environment.
Many carcinoma cells, however, lose such polarized phenotypes during tumor progression. Little is known about the
mechanism of how the polarized epithelial cell-cell contacts
are established, or of how they are disrupted during carcinogenesis.
C
polarity typical of simple epithelia; they formed
microvilli only on the outer surface of the aggregates,
and a junctional complex composed of tight junction,
adherens junction, and desmosome arranged in this
order. These results indicate that the activation of
E-cadherin triggered the formation of the junctional
complex and the polarized distribution of cell surface
proteins and structures. We also found that, in untransfected PC9 cells, ZO-1 formed condensed clusters
and colocalized with E-cadherin, but that other adhesion molecules rarely showed such colocalization with
E-cadherin, suggesting that there is some specific interaction between ZO-1 and E-cadherin even in the absence of cell-cell contacts. In addition, we found that
the activation of E-cadherin caused a retardation of
PC9 cell growth. Thus, we concluded that the
E-cadherin-catenin adhesion system is essential not
only for structural organization of epithelial cells but
also for the control of their growth.
It has been suggested that the cadherin cell-cell adhesion
system plays a role in initiation of polarized cell-ceU association (Gumbiner and Simons, 1986; Gumbiner et al., 1988).
Cadherins are a family of cell-cell adhesion molecules
(Takeichi, 1991). A subfamily of them, the so called"classic"
cadherins including E-cadherin, are localized in the adherens junctions (AJ) 1 (Kemler, 1993), and another subfamily, composed of desmogleins and desmocoUins, are
restricted to desmosomes (Buxton et al., 1993). Without
cadherins, cells are unable to tightly associate with each
other, and do not show polarity. Introducing a cadherin into
cadherin-deficieut cells induces a polarized distribution of
surface proteins such as Na+, K+-ATPase (McNeill et al.,
1990) and enhances gap junction formation (Mege et al.,
1988; Matsuzaki et al., 1990). On the other hand, blocking
cadherin function in epithelial cells with antibodies results
in disruption of the junctional complexes (Gumbiner and Si-
Address all correspondence to Masatoshi Takeichi, Department of Biophysics, Faculty of Science, Kyoto University, Kitashir kaw , Sakyo-ku,
Kyoto 606-11, Japan. Ph.: (81) 75 751-2111. Fax: (81) 75 753 4197.
1. Abbreviation used in this paper: AJ, adherens junctions.
© The Rockefeller University Press, 0021-9525/94/10/247/10 $2.00
The Journal of Cell Biology, Volume 127, Number l, October 1994 247-256
247
Downloaded from on October 2, 2016
sociate with one another, and therefore grow singly,
despite their expression of E-cadherin, because of
their lack of ot-catenin, a cadherin-associated protein.
However, when the E-cadherin is activated by transfection with a-catenin cDNA, they form spherical aggregates, each consisting of an enclosed monolayer
cell sheet. In the present work, we examined whether
the a-catenin-transfected cell layers expressed epithelial phenotypes, by determining the distribution of various cell adhesion molecules on their surfaces, including E-cadherin, ZO-1, desmoplakin, integrins, and
laminin. In untransfected PC9 cells, all these molecules were randomly distributed on their cell surface.
In the transfected cells, however, each of them was
redistributed into a characteristic polarized pattern
without a change in the amount of expression. Electron microscopic study demonstrated that the
ot-catenin-transfected cell layers acquired apical-basal
Published October 1, 1994
Materials and Methods
Cell Cultures and aE-Catenin cDNA Transfection
Human lung carcinoma PC9 (Kinjo et al., 1979) and PC9-c~NA, a line of
PC9 transfected with chicken ¢xN-catenin eDNA (Hirano et al., 1992), were
cultured in a 1:1 mixture of DME and Ham's F12 supplemented with 10%
FCS (DHI0).
The expression vector of mouse c~E-catenin cDNA, pBAT-c~,was made
by replacing the BglII-XbaI fragment of pBAT-EM2 (Nose et al., 1988) with
the BamHI-XbaI fragment of pSK102B (Nagafuchi and Tsukita, 1994).
Transfection of PC9 cells with pBAT-~ was performed as described previously (Hirano et al., 1992). Cells were cotransfected with pPGKneo bpA
(Soriano et al., 1991) and selected in G418-containing medium. Isolated
G-418-resistant cell clones were tested for their expression of c~E-catenin by
immunoblotting.
Antibodies
The following antibodies were used: mouse mAbs HECDol (Shimoyama
et ai., 1989) and SHE78-7 (Ihkara Shuzo Co., Ltd., K3~to, Japan) to E-cadherin; rat mAb ECCD-2 to E-cadherin (Shirayoshi et ai., 1986); rat mAb
1809 to c~E-catenin (Nagafuchi and Tsukita, 1984); rabbit polyclonal anti-
The Journal of Cell Biology, Volume 127, 1994
body (pAb) to/3-catenin (Shibamoto et al., 1994); mouse mAb PG5.1 to
plakoglobin (61005; Progen Biotechnik GmbH, Heidelberg, Germany);
mouse mAb "1'8-754 to ZO-I (Itoh et ai., 1991); mouse mAb 11-5F to desmoplakin (Parrish et ai., 1987); rabbit pAb 4316 to 131 integrin (a gift of
Kenneth Yamada); rat mAb 13 to/31 integrin (Akiyama et al., 1989); and
rabbit pAb to laminin (AT-2404, E-Y Laboratories, Inc., San Mateo, CA).
For detection of primary antibodies, we used sheep biotinylated speciesspecific antibody to mouse Ig (RPN1001; Amersbam International, Amersham, UK) and to rat Ig (RPN1002; Amersham International); donkey biotinylated species-specific antibody to rabbit Ig (RPNI004; Amersham
International); FITC-labeled streptavidin (RPNI232; Amersham International); Texas red-labeled streptavidin (RPN1233; Amersham International); goat Cy3-1abeled species-specific antibody to rat IgG (AP-183C;
Chemicon International, Inc., Temucula, CA); sheep FITC-labeled antibody to mouse Ig (N1031; Amersham International); goat rbodaminelabeled antibody to rat IgG (55763; Organon Technika N. V.-Cappel Products, West Chester, PA); sheep HRP-linked species-specific antibody to
mouse Ig (NA9310; Amersham International) and to rat Ig (NA932; Amersham International); goat HRP-linked antibody to rabbit IgG (3212-0081;
Cappel Products); and Sepharose 4B-linked goat antibody to mouse IgG
(62-6541; Zymed Laboratories, Inc., San Francisco, CA), and to rat IgG
(62-9541; Zymed Laboratories).
Immunofluorescence Staining
Cells suspended in culture medium were collected by centrifugation,
resuspended in Hepes-buffered (pH 7.4) HBSS, and placed on coverslips
that had been pre-coated with 1 mg/wi polyethylenediamine by incubation
overnight at room temperature. After 15 rain at room temperature, the cells
were fixed as follows. For immunofluorescence staining for desmoplakin,
cells were fixed and permeabilized by treatment with methanol on ice for
5 win. For staining for other molecules, cells were treated as for the desmoplakin staining, or they were fixed with 3.5 % paraformaidehyde in HBSS
at 4°C for 20 min, and permeabilized by treatment with methanol at - 2 0 ° C
for 20 win. For avoidance of nonspecific binding of antibodies, the fixed
cells were pre-treated with 5% skim milk for more than 15 win. They were
subsequently incubated with a primary antibody, a biotinylated secondary
antibody, and FITC or Texas red-labeled streptavidin. For doubleoimmunostaining, fixed cells were first treated with a mixture of two primary antibodies, followed by incubation with a biotinylated antibody to one primary antibody, and then a mixture of FTI'C or Texas red-labeled streptavidin and
Texas red- or FITC-labeled secondary antibody to the other primary antibody. In some experiments, Cy3-1abeled secondary antibody was substituted for Texas red-labeled antibody. Cross-reactivity of each secondary
antibody sample was carefully checked to avoid false double-staining prior
to experiments. The immunostained samples were examined under Zeiss
Axiophot and confocai microscopes. Confocal pictures were taken to show
single optical x-y sections.
lmmunoblotting
Cells were scraped out of dishes with a rubber policeman, dissolved in
Laemwii's SDS sample buffer, and boiled with 5 % ~-mercaptocthanol for
10 win. Proteins were separated by 7.5, 5, or 3.5% polyacrylamide gel electrophoresis, followed by transference to nitrocellulose filters. The filters
were incubated with primary antibodies and then with HRP-conjugated secondary antibodies. Bound HRP was visualized by use of the enhanced chemiluminescence system (RPN2106, Amersham International).
Immunoprecipitation
All steps in the following immunoprecipitation protocol were carried out
either on ice or at 4°C. Cells attached to dishes were scraped off with a rubher policeman, and the resulting suspension of cells was transferred to a test
tube. The cell suspensions were centrifuged at 150 g for 5 win, and the
pellets were resnspended in HBSS and transferred to Eppendorf tubes. Cells
were centrifuged, and then 1 wi of an ice-cotd detergent extraction buffer
(1% NP-40, 1% Triton X-100, 1 mM CaCI2, and 1 mM PMSF in 50 mM
TBS, pH 7.6) was added to each pellet. After incubation for 30 win, the
crude extracts were centrifuged at 100,000 g for 10 n~n. The supernatants
were transferred to Eppendorf tubes. The extracts were precleared by incubation with Sepharose 4B for 30 rain. These samples were then centrifuged
at 1,200 g for 10 s, and the resulting supernatants were transferred to new
tubes. The primary antibody was added to the cleared extr~ts, which were
incubated for 1 h. Then the extracts were incubated with Sepharose 4B-
248
Downloaded from on October 2, 2016
mons, 1986; Gumbiner et al., 1988; Meyer et al., 1992),
suggesting that cadherins are prerequisite components for
the formation of these structures. Recent studies suggest that
E-cadherin-mediated adhesion might activate a PKC signaling system to induce the formation of tight junction (Balda
et al., 1993).
Our recent studies presented another example showing the
importance of cadherin function in epithelial organization
(Hirano et al., 1992). Cadherins in the AJ are associated
with a group of cytoplasmic proteins, o~, ~-catenins (McCrea
et al., 1991) and plakoglobin (Knudsen and Wheelock,
1992; Piepenhagen and Nelson, 1993). ot-Catenin is divided
into two subtypes, otE-catenin, which is expressed in many
kinds of tissues (Nagafuchi et al., 1991; Herrenknecht et al.,
1991; Nagafuchi and Tsukita, 1994), and c~N-catenin, expressed mainly in the nervous system (Hirano et al., 1992;
Uchida et al., 1994). Cells of the lung carcinoma line PC9
express neither of these ot-catenin subtypes but do express
E-cadherin and ~/-catenin, and they cannot aggregate and
thus grow singly in suspension (Hirano et al., 1992; Shimoyama et al., 1992). Transfection of PC9 cells with otN-catenin cDNA activates the E-cadherin, leading them to acquire
cadherin-mediated cell-cell associations (Hirano et al.,
1992). Moreover, these transfected cells display an epithelioid arrangement in the aggregates. These results indicate
that not only is c~-catenin indispensable for cadherin function
but also that the cadherin-catenin system provides some signals for cells to arrange themselves into an epithelioid
pattern.
In the present study, we performed detailed analyses of the
structure of the ot-catenin-transfected PC9 cells and also of
their growth property. The results show that these cells acquired apical-basal polarity including the formation of a
junctional complex, which is characteristic of simple epithelia. Thus, the dispersed carcinoma cells regained an almost
complete set of epithelial-specific adhesive properties as the
result of c~-catenin transfection. We also show that the growth
of PC9 cells is significantly suppressed following ot-catenin
transfection. These results suggest that the cadherin-mediated cell-cell contacts regulate not only the structural organization of cell layers but also their growth property.
Published October 1, 1994
linked secondary antibody for 30 rain and centrifuged at 1,200 g for 10 s.
The pellets were rinsed three times with the extraction buffer, and resuspended in Laemmirs SDS sample buffer and boiled with 5 % ~-mereaptoethanol for 10 rain. The samples were centrifuged at 3,500 g for 10 min,
and the supernatants were used for immunoblotting.
Measurement of Cell Growth
Figure1. Morphology of parental PC9 and PC9-c~EIcells. (a) PC9;
(b-d) PC9-c~E1.Cells in d were cultured with 10 #g/ml of SHE78-7
for 4 d. Living (a, b, and d) or fixed (c) cells were photographed
under phase-contrast optics. Bars: (a, b, d) 100 #m; (c) 40 #m.
Results
Adhesive Properties of PC9 Cells Transfected with
aE- and ¢tN-Catenin cDNA
separated and cultured them independently. Each of them,
however, always produced both nonattached and attached
cells in a culture, suggesting that these two forms are convertible into each other. Probably, the aggregated cells occasionally attach to the substratum and spread on it, but some
of them again detach and grow in the form of aggregates. On
the other hand, the majority of the untransfected PC9 cells
remained in suspension except for a certain population of
attached cells. In the following experiments, the two cell
populations were not separated for biochemical and growth
analyses, but the suspended form was mainly used for morphologicai analyses. In addition, a-catenin expression in the
transfected lines was not extremely stable, and nonadhesive
single cells were constantly produced in cultures of the transfectants. Thus, a certain level of contamination of those cells
was unavoidable in the assay of the transfectants, but their
presence did not largely affect the results.
In our previous work, PC9 cells were transfected with
chicken aN-catenin cDNA, and a transfectant line, PC9aNA, was isolated (Hirano et al., 1992). Since the PC9 line
was derived from lung carcinoma (Kinjo et al., 1979), the
original a-catenin of the cells was supposedly aE-catenin.
To restore aE-catenin expression to this cell line, we transfected PC9 cells with mouse aE-catenin cDNA and obtained
transfectant lines, one of which was designated as PC9-aE1.
Cells of the aE-catenin-transfected lines formed aggregates
(Fig. 1 b), each consisting of an enclosed epithelioid cell
monolayer (Fig. 1 c), in contrast to the dispersed nature of
the parent PC9 cells (Fig. 1 a). When PC9-aE1 cells were
cultured with anti-E-cadherin antibodies, they did not form
such epithelioid aggregates, being indistinguishable from the
untransfected PC9 cells in morphology (Fig. 1 d), as previously found for PC9-aNA. Thus, both subtypes of a-catenin
showed a similar effect on E-cadherin-dependent aggregation of PC9 cells. In the following experiments, PC9-aE1
was generally used, and PC9-aNA was occasionally used to
confirm the generality of the phenotypes of a-catenin-transfected PC9 cell lines. Both cell lines gave essentially the
same results in all the following experiments. The term
a-catenin was used as a collective name of aE- and aNcatenin.
We noticed that the cultures of the above transfectant lines
contained not only spherical aggregates but also a cell population adherent to the dish (Fig. 1 b). The latter formed
monolayer cell sheets. The proportion of the attached and
nonattached cells varied from culture to culture. To determine the relationship between the two cell populations, we
Simple epithelia are characterized by polarized distributions
of cell adhesion structures and molecules. To determine
whether the epithelioid PC9-aE1 cell layers indeed express
epithelial phenotypes, we studied the distribution of several
adhesion-related proteins, ZO-1 (tight junction protein),
E-cadherin,/31, a2, a3, and a6-integrins, desmoplakin (desmosomal plaque protein), and laminin (basement membrane
protein) by conventional or confocai immunofluorescence
microscopy. In untransfected PC9 cells, all these molecules
were randomly distributed on the cell surface or in the cytoplasm. ZO-1 was detected as a limited number of small clusters with an irregular shape on the surface of the cells (Fig.
2 a). E-cadherin (Fig. 2 b),/~l-integrin (Fig. 2 c), and lami-
Watabe et al. Cadherin-Catenin in Epithelial Organization
249
Formation of Polarized CeU-CeU Association after
a-Catenin cDNA Transfection
Downloaded from on October 2, 2016
For preparation of cell cultures for measuring growth, a,-catenin-transfected PC9 cells forming compact aggregates were dissociated by treatment
with 0.05% trypsin (Difco Laboratories, Detroit, MI) for 30 rain at 37"C,
and their number was counted. Untransfeeted PC9 cells were treated similarly, although this process was not necessary for the purpose of cell counting because of their dispersed nature. One million of the trypsinized cells
were plated in a 100-ram dish and cultured for 1 d. Then, the same trypsin
treatment was performed again to obtain a completely dissociated cell suspension, which was necessary for accurate cell eotmting and also for starting the transfectant cultures under the same conditions as for the parent
cells. 1,000 of these cells were re-plated with 0.5 ml of DH10 in each well
of a 24-well plate (A/S Nunc, Roskilde, Denmark), and cultured with or
without purified HECD-1 or SHE78-7 antibodies. After 3 d 0.5 ml fresh
medium of the same type was added. Then haif of the medium was changed
every three days, to which fresh antibodies were added each time from the
stock solutions. For determination of cell proliferation, cells unattached in
each well were collected into a tube, centrifuged, rinsed once with HBSS
without Ca :+ and Mg 2+ (HCMF) but with 1 mM EDTA, and suspended in
the same solution. Then, this cell suspension was returned to the original
well that contained cells attached to the bottom. To this sample, trypsin was
added to the final concentration of 0.05 %, and the plate was incubated for
30 min at 37"C. The number of dissociated cells were counted with a
hemocytometer.
Published October 1, 1994
Figure2. Immunofluorescence staining for junctional proteins on
nin (data not shown) were scattered on the cell surface or in
the cytoplasm as diffused clusters. The distributions of
tx-integrins were similar to the distribution of the ill-chain
(data not shown). Desmoplakin was detected mostly in the
cytoplasm as small clusters (Fig. 2 d), as found in low Ca 2÷
media (Mattey and Garrod, 1986b; Green et al., 1987; Pasdar and Nelson, 1988).
Interestingly, when ZO-1 and E-cadherin were doublestained, some ZO-1 clusters, but not all of them, were
colocalized with E-cadherin staining (compare Fig. 2, a and
b); this was observed under two different fixation conditions,
methanol treatment alone, and paraformaldehyde fixation
followed by methanol treatment. The ZO-1 clustering itself
occurred without E-cadherin, since a similar ZO-1 distribution was observed in a population of PC9 cells not expressing
E-cadherin, which cells were often observed in the PC9 cultures. On the other hand, the majority of E-cadherin sites
were free of ZO-I. Bl-integrin and desmoplakin did not show
such an obvious colocalization with E-cadherin on the cell
surface, although their colocalization was sometimes observed in the cytoplasm.
In aggregates of PC9-otE1 cells, all the above adhesion
molecules exhibited polarized distributions. The outer surface of the aggregates was devoid of any of these molecules.
ZO-1 was detected in a pinpoint immunofluorescence at the
outermost portion of cell-cell contacts (Fig. 3 a), as found
in normal simple epithelial cells. The ZO-1 staining was contiguous to E-cadherin staining, as observed in doubleimmunostained samples (Fig. 3, a and b). E-cadherin was
localized to lateral cell-cell contacts (Fig. 3 b), but it tended
to be concentrated more in cell-cell contact in the outer
regions of many aggregates (Fig. 3 d). On the other hand,
The Journalof Cell Biology,Volume127, 1994
Expression of Cell Adhesion Molecules before and
after ~-Catenin Transfection
We tested whether the expression level of the various cell
adhesion molecules examined was changed or not after
ot-catenin transfection. Immunoblot and immunoprecipitation analyses showed that there was little difference in the
level of ZO-I, laminin, desmoplakin,/~l-integrin (Fig. 5), or
E-cadherin (Fig. 6, lanes 5 and 6) before and after the ot-catenin transfection. These results suggest that cell-cell junctions formed in the oL-catenin transfected cells were organized without additional synthesis of necessary components.
We then examined by immunoprecipitation experiments
whether the expression of o~-catenin affected the association
of E-cadherin with other catenins. In detergent extracts of
PC9 cells, E-cadherin coimmnnoprecipitated with i-catenin
and plakoglobin (Fig. 6, lanes 1 and 3). In PC9-o~EI cells,
similar amounts of i-catenin or plakoglobin were coprecipitated with E-cadherin (Fig. 6, lanes 2 and 4). This suggests
that the binding of these two proteins to E-cadherin is not
affected by o~-catenin. We also tested the possibility that
ZO-I and E-cadherin may be co-purified upon immunoprecipitation, because of their colocalization in PC9 cells, but
the results were negative (data not shown).
250
Downloaded from on October 2, 2016
parental PC9 cells. Fixed cells were double-immunostained for
ZO°I (a) and E-cadherin (b), or stained for Bl-integrin (c) or desmoplakin (d). Note that ZO-1 colocalizes with E-cadherin. Bar,
I0 ~m.
fll-integdn was detected evenly at cell-cell contacts as well
as at the inner free surface of the aggregates; obvious differences between the fll-integrin and E-cadherin distributions
were observed in double-immunostained samples (compare
Fig. 5, c and d). All a-chains of the integrin studied here
colocalized with ill-chain (data not shown). Desmoplakin
was detected as characteristic punctuated lines at cell-cell
boundaries (Fig. 3 e), and laminln was localized in both
cell-cell contacts and the inner surfaces of the aggregates,
whose pattern was similar to that of integrins (Fig. 3 f ) . The
cellular distributions of all these proteins are quite similar
to those in normal simple epithelia, except for laminin; this
protein is not found in cell-cell contacts in normal tissues.
These results suggest that the txE-catenin-transfected PC9
cell layers acquired apical-basal polarity, in which the apical
pole was oriented to the outer surface of the aggregates.
PC9-txNA cells showed a similar phenotype with regard to
the distribution of these cell adhesion molecules. The polarized distribution of cell adhesion molecules in these ot-catenin-transfected PC9 cells was abolished when the ceils were
cultured in the presence of anti-E-cadherin antibodies (data
not shown).
To confirm the above observations, we used electron microscopy. Untransfected PC9 cells did not exhibit any polarized and specialized structures on their surface (data not
shown). In contrast, PC9-txE1 aggregatesshowed various
polarized structures: in low-power views, we could observe
microvilli only on the outer surface of the aggregates (Fig.
4 a), which was reminiscent of the apical surface of normal
simple epithelia. Closer examinations, then, demonstrated
that tight junction, adherens junction and desmosome were
present in this order from the outside in many aggregates
(Fig. 4 b). This set of junctional structures was indistinguishable from the typical junctional complex observed in normal
simple epithelia (Farquhar and Palade, 1963). However, we
could not detect the basement membrane in PC9-txE1 aggregates.
Published October 1, 1994
Further, we tested the detergent solubility of E-cadherin
before and after ot-catenin transfection. Most of E-cadherin
molecules were solubilized with the detergent extraction
buffer containing 1% NP-40 and 1% Triton-X in the absence
of ot-catenin (Fig. 7, A and B, lanes 1 and 2). In PC9-otE1
cells, however, an insoluble fraction of E-cadherin was detected after extraction with the same detergents for 5 to 10
min (Fig. 7 A, lane 4), although this fraction disappeared after a 1-h extraction of the cells (Fig. 7, lane 4). Thus, E-cadherin was more resistant to detergent extraction after ct-catenin transfection.
Retarded Growth of the ct-Catenin-transfected
PC9 Cells
cells is directly associated with the activation of E-cadherin,
the effect of blocking antibodies to E-cadherin was examined. By the antibody treatments, the growth of PC9-otE1
cells was considerably enhanced (Fig. 8); two blocking
mAbs HECD-1 and SHE78-7 showed a similar effect, but
other antibodies, that could not disrupt cell-cell contacts,
had no effect. The growth rate of the transfectants in the presence of anti-E-cadherin antibodies was a little lower than
that of the parent PC9 cells even in the presence of saturated
concentrations of the antibodies. This is probably due to either a possible failure in the complete blocking of E-cadherin
by the antibodies or a clonal variation in growth rate between
the parent and transfectant lines.
Finally, we studied whether or not proliferation of PC9 cells
was affected by c~-catenin transfection. As shown in Fig. 8,
PC9-txE1 cells grew much more slowly than the parent PC9
cells. To check whether this growth retardation of PC9-otE1
Two important findings were made in this work. First, the
activation of E-cadherin by means of t~-catenin transfection
Watabe et al. Cadherin-Catenin in Epithelial Organization
251
Discussion
Downloaded from on October 2, 2016
Figure 3. Confocal images of immunofluorescence staining for adhesion proteins on PC9-aE1 cells. (a
and b) Double-staining for ZO-1 (a)
and E-cadherin (b), (c and d) Double-staining for /~l-integrin (c) and
E-cadherin (d). (e) Desmoplakin;
(f) laminin.
Published October 1, 1994
of the originally dispersed PC9 carcinoma cells induced
polarized cell-cell associations characteristic of simple epithelia. Second, the restoration of E-cadherin-mediated
cell-cell contacts in this cell line caused growth retardation.
These effects of o~-catenin transfection were abolished by
treatment with antibodies specific for E-cadherin, indicating
that E-cadherin and t~-catenin are not separable in their functions. The phenomena observed here should thus be interpreted to be based on a cooperative action of these two
molecules, not on the sole action of ot-catenin which was exogenously introduced.
The role of t~-catenin in cadherin activity is not fully understood. The present study demonstrated that c~-catenin
transfection of PC9 cells slightly reduced the detergent solubility of E-cadherin. This finding supports the idea that
c~-catenin may play a role in mediating the interactions of
cadherins with the actin-based cytoskeleton (Ozawa et al.,
The Journal of Cell Biology,Volume 127, 1994
1990; Kemler, 1993). In addition, we found that binding of
fl-catenin or plakoglobin to cadherins was independent of the
presence or absence of tx-catenin, as consistent with previous
observations (McCrea and Gumbiner, 1991).
Previous studies have suggested that the classic cadherinmediated cell-cell contact is a prerequisite for the formation
of tight junction and desmosome. This idea is supported by
indirect evidence such as that tight junction formation is
Ca 2+dependent (see reviews by Citi, 1993; Anderson et al.,
1993) and that the desmosome is sensitive to Ca ~+ as well
(Volk and Geiger, 1986; Pasdar and Nelson, 1986; Mattey
and Garrod, 1986a, b; Green et al., 1987; Gumbiner et al.,
1988). More direct evidence was obtained by the observation
that antibodies to E-cadherin (uvomorulin) inhibited the formation of these junctions (Gumbiner and Simons, 1986;
Gumbiner et al., 1988; Fleming et al., 1989; Balda et al.,
1993). However, such evidence is not conclusive for the role
252
Downloaded from on October 2, 2016
Figure 4. Electron microscopic observation of a cross section of a PC9-c~E1 cell
aggregate. (a) A low-power view. (b) An
enlargement of the portion indicated by
arrowhead in a. Microvilli are observed
only on the outer surface of the structure,
and tight junction (TJ), adherens junction
(AJ), and desmosome (DS) are observed
at cell-cell contacts. Insert in b enlarges
the tight junction in the same figure. Bars:
(a) 2 #m; (b) 200 nm.
Published October 1, 1994
Figure 5. Immunoblot analysis of the expression of adhesion proteins in PC9 (lane 1) and PC9-c~E1 (lane 2) cells. (A) ZO-1. (B)
Laminin. BI- and B2-chain. (C) Two isoforms of desmoplakin.
Other bands are perhaps degradation products. (D) Bl-integrin. The
specific protein bands detected are shown by arrowheads. Ig
represents immunoglobulin chains. Whole cell lysates (A and B) or
immunoprecipitants obtained with anti-desmoplakin antibody (C)
or anti-/~l-integrin antibody (D) were subjected to immunoblotting.
Polyacrylamide gels of 3.75, 5, and 7.5% were used for A and B,
C, and D, respectively. Positions of molecular weight markers are
200, 116, and 97 x 103 for A and B; 200, 116, 97, and 66 x 103
for C; and 116, 97, 66, and 45 x 103 for D.
Thus, the present study provided the most clear-cut evidence
for the requirement of the classic cadherin adhesion system
for the formation of the junctional complex.
How can E-cadherin initiate the formation of other junctions? Untransfected PC9 cells expressed all the adhesion
molecules studied, and their expression level was not significantly altered following ot-catenin transfection. E-cadherin-mediated cell-cell contacts, therefore, seem to have
6-
~-" 5"
0
.0
E
c-
"~O
4
-~o--
PC9
PC9 + 10A
3
.
0
.
.
.
,
.
.
.
.
5
i
10
.
•
ct
_•
ct+A
=
ct+ 10A
.
.
.
15
Figure 6. Immunoblot analysis of E-cadherin-associated proteins.
An equal number of PC9 cells (lanes 1, 3, 5, and 7) and PC9-t~E1
cells (lanes 2, 4, 6, and 8) were detergent-extracted and -soluble
proteins were subjected to immunoprecipitation with HECD-1. Immunoprecipitants were detected by immunoblotting with antibodies
to ~-catenin (lanes 1 and 2), to plakoglobin (lanes 3 and 4), to
E-cadherin (lanes 5 and 6), and to u-catenin (lanes 7and 8). Note
that almost equal amounts of/3-catenin and plakoglobin relative to
E-cadherin were coprecipitated with the E-cadherin, regardless of
the binding of c~-catenin. Positions of molecular weight markers are
200, 116, 97, 66, and 45 x 103.
Figure 8. Effect of ot-catenin transfection and anti-E-cadherin antibody treatment on cell growth. Equal numbers of PC9 cells (PC9)
and PC9-ctE1 cells (c0 were seeded in wells, and the cell numbers
were counted 6 and 12 d after incubation without or with 0.1 #g/ml
(+ A) or 1 #g/ml (+ 10,4) HECD-1. The former concentration was
sufficient to block E-cadherin-mediated aggregation of PC9-otE1
cells. The horizontal axis depicts the days after seeding, and the
vertical axis common logarithms of cell numbers. Vertical error
bars represent standard deviations.
Watabe et al. Cadherin-Catenin in Epithelial Organization
253
days
Downloaded from on October 2, 2016
of cadherins in junctional complex formation, because the
effect of anti-cadherin antibodies was always only partial in
inhibition. Most anti-cadherin antibodies cannot completely
block cadherin activity in a given cell; one reason for this
inability is that the majority of cells express multiple types
of cadherin including unidentified ones, and another reason
is that there is a limit in the action of antibodies in blocking
cadherin activity, especially when monoclonal antibodies
are used. In the present approach, we could overcome these
problems by use of a cell line in which normal cadherin activity was genetically blocked (Shimoyama et al., 1992).
Figure 7. Detergent extractability of E-cadherin. PC9 cells (lanes
1 and 2) and PC9-ctE1 cells (lanes 3 and 4) were extracted with
a mixture of NP-40 and Triton X-100 for I0 min (,4) or 1 h (B),
and soluble (lanes I and 3) and insoluble fractions (lanes 2 and 4)
were subjected to immunoblotting with HECD-1. With the 10-min
treatment, all E-cadherin molecules in PC9 cells were extracted,
while a small fraction of insoluble E-cadherin molecules were detected in PC9-,E1 cells. Positions of molecular weight markers are
200, 116, 97, 66, and 45 x 103.
Published October 1, 1994
The Journal of Cell Biology, Volume 127, 1994
1992), but opposite to that in normal epithelia in which the
apical surface faces the lumen enclosed by a cell sheet. A
similar inverted polarity is often observed with aggregates of
other epithelial cells in vitro (Rodriguez-Boulan and Nelson,
1989), and the orientation of cell polarity seems to depend
on cellular environments or intrinsic phenotypes. The only
abnormal features in the ot-catenin-transfected PC9 cells
were the lack of the basement membrane and the unusual distribution of laminin, which was localized all over the
basolateral surfaces of these ceils. The reason for this
phenomenon remains to be studied; the formation of the
basement membrane perhaps requires some factors that PC9
cells fail to express.
Concerning the inhibitory effect on cell growth of the activation of E-cadherin, a similar phenomenon was reported
based on studies using certain carcinoma lines (Navarro et
al., 1991), but such inhibition is not widely seen in other systems. Probably, sensitivity of cell growth to cadherin-mediated cell-cell contacts varies with the cell types examined.
Actually, many tumor cells are highly proliferative regardless of their normal cadherin activity and tight cell-cell associations, suggesting their inability to respond to cell contacts. However, the growth of normal cells is known to be
contact dependent. From this point of view, it is likely that
PC9 cells might have responded to the E-cadherin-mediated
cell contacts in a way of normal cells. As to the mechanism
of how cadherin-mediated cell-cell contacts affect cell
growth, at least four possibilities can be considered. First,
cadherin-cadherin interactions at their extracellular domain
may generate some signals to regulate cell growth, in a manner as found in a variety of receptor-ligand interactions. The
AJ contains molecules that could serve for signaling
processes, e.g., members of the Src tyrosine kinase family
are localized in this cell junction (Tsukita et al., 1991; Tsukita et al., 1993). Second, cell-cell association mediated by
cadherins may trigger interactions of other cell surface
receptors with their ligands, leading to growth control. If the
ligands are membrane-bound, cadherin-mediated cell-cell
contact may be a prerequisite for their interactions with the
receptors. It is known that some receptors such as EGF
receptors are localized around the AJ (Fukuyama and
Shimizu, 1991). Third, cadherins may control cell growth
via the induction of tight junction formation. ZO-1, a component of the tight junction, has homology to the product of
the Drosophila lethal(1 )discs-large (dlg) gene (Woods and
Bryant, 1991; Itoh et al., 1993; Willott et al., 1993), which
mutation causes abnormal overgrowth of tissues (Woods and
Bryant, 1991). Tight junction formation thus could affect cell
growth. Fourth, the layers of cells connected with the junctional complex could be less accessible to growth factors in
culture medium than dispersed cells. This possibility cannot
be ignored when essential growth factor receptors are localized to the basolateral surfaces, especially for closed cystic
aggregates in which tight junctions could form a perfect permeability barrier to extracellular factors. Whatever the mechanism is, our present findings indicate that the cadherincatenin adhesion system may serve as a direct or indirect
regulator of cell growth.
In conjunction with the above discussion, an intriguing observation was recently reported, that the tumor suppressor
gene APC product binds to/~- and t~-catenins (Rubinfeld et
al., 1993; Suet al., 1993), implying that catenins might be
254
Downloaded from on October 2, 2016
merely induced their structural reorganization. The simplest
model for the role of E-cadherin would be that this molecule
plays no specific roles in the formation of other junctions, except providing initial cell-cell contacts. Tight junctions and
desmosomes could be formed regardless of cadherin function, if two cell membranes were brought into close contacts
with each other. On the other hand, it is equally possible that
E-cadherin plays a signaling role in the formation of other
junctions. A recent study suggested that PKC activation is involved in tight junction formation, and this process may be
initiated by E-cadherin-mediated cell contacts (Balda et
al., 1993).
We should also consider the possibility that there may be
structural links between the E-cadherin-catenin complex
and other junctions, and the former may control the assembly of the latter through this system, including their polarized arrangement. In this context, it should be particularly
noted that E-cadherin was found to colocalize with ZO-1 on
the free surface of PC9 cells, as observed in certain types
of cell-cell junctions (Itoh et al., 1993). An important aspect
of this finding is that this colocalization occurred between
free molecules in the absence of cell-cell contacts. This suggests that these two molecules can interact with one another,
directly or indirectly, at the prejunctional stage. Tight junctions and the cadherin-based AJs are formed side by side in
most epithelial cells. This may not be coincidental, and the
interaction between cadherin and ZO-1 may serve to maintain their intimate spatial relationship. Since we could not
copurify ZO-1 and E-cadherin by immunoprecipitation,
their interaction, if present, may be not so stable. On the
other hand, the formation of desmosomes could be more or
less independent of classic cadherin function, as antibodies
to classic cadherins do not always disrupt desmosomes once
formed (Wheelock and Jensen, 1992; Watabe et al., 1993;
Hodivala and Watt, 1994). The present observation that PC9
cells do not form desmosomes without active E-cadherin,
however, suggests that desmosomal formation requires at
least preceding cell-cell contacts.
We found in the present study that activation of E-cadherin
was sufficient to reconstruct an almost complete epitheliumspecific architecture in cells which otherwise would have
remained disperse. The restored phenotypes included the
formation of the junctional complex, the basolateral localization of integrins, and the apical development of microvilli,
all of which are characteristic of normal simple epithelia.
Previous studies showed that introduction of cadherin cDNA
into cells that have no endogenous cadherins was capable of
redistributing cell surface proteins into a polarized pattern.
In those studies, however, only partial epithelium-specific
phenotypes were restored in the transfected cells; for example, L cells transfected with E-cadherin cDNA did not form
tight junctions (McNeill et al., 1990). Differences between
the present and previous results may be attributed to differences in the cell lines used. PC9 cells are supposed to be of
epithelial origin, and must genetically inherit most of their
original phenotypes; reactivation of the cadherin system
must have triggered the re-expression of these phenotypes.
Cell lines like L, which cannot reorganize epithelium-specific structures, may not be of epithelial origin or may have
lost genes necessary for epithelial phenotypes. The apicalbasal polarity formed in the ot-catenin-transfected PC9 cells
was similar to that in mouse blastocysts (Fleming et al.,
Published October 1, 1994
Akiyama, S. K., S. S. Yamada, W.-T. Chen, andK. M. Yamada. 1989. Analysis of fibronectin receptor function with monoclonal antibodies: roles in cell
adhesion, migration, matrix assembly, and cytoskeletal organization. J. Cell
Biol. 109:863-875.
Anderson, J. M., M. S. Balda, and A. S. Fanning. 1993. The structure and
regulation of tight junctions. Curr. Opin. Cell Biol. 5:772-778.
Balda, M. S., L. Gonzalez-Mariscal, K. Matter, M. Cereijido, and J. M. Anderson. 1993. Assembly of the tight junction: the role of diacylglycernl. J.
Cell Biol. 123:293-302.
Behrens, J., L. Vakaet, R. Friis, E. Winterhager, F. Van Roy, M. M. Marcel,
and W. Birchmeier. 1993. Loss of epithelial differentiation and gain of invasiveness correlates with tyrosine phosphorylation of the E-eadherin//~-catenin complex in cells transformed with a temperature-sensitive v-SRC gene.
J. Cell Biol. 120:757-766.
Burdsal, C. A., C. H. Damsky, and R. A. Pedersen. 1993. The role of E-cadherin and integrins in mesoderm differentiation and migration at the mammalian primitive streak. Development. 118:829-844.
Buxton, R. S., P. Cowin, W. W. Franke, D. R. Garrod, K. J. Green, I. A.
King, P. J. Koch, A. I. Magee, D. A. Rees, J. R. Stanley, and M, S. Steinberg. 1993. Nomenclature of the desmosomal cadherins. J. Cell Biol. 121 :
481-483.
Citi, S. 1993. The molecular organization of tight junctions. J. Cell Biol.
121:485-489.
Farquhar, M. G., and G. E. Palade. 1963. Junctional complexes in various
epithelia. J. Cell Biol. 17:375-412.
Fleming, T. P., J. McConnell, M. H. Johnson, and B. R. Stevenson. 1989. De-
velopment of tight junctions de novo in the mouse early embryo: control of
assembly of the tight junction-specific protein, ZO-1. J. Cell Biol. 108:
1407-1418.
Fleming, T. P., Q. Javed, and M. Hay. 1992. Epithelial differentiation and intercellular junction formation in the mouse early embryo. Development.
Suppl. 105-112.
Fukuyama, R., and N. Shimizu. 1991. Detection of epidermal growth factor
receptors and E-cadherins in the basolateral membrane of A431 cells by laser
scanning fluorescence microscopy. Jpn. J. Cancer Res. 82:8-11.
Green, K. J., B. G-eiger, J. C. R. Jones, J. C. Talian, andR. D. Goldman. 1987.
The relationship between intermediate filaments and microfilaments before
and during the formation of desmosomes and adherens-type junctions in
mouse epidermal keratinocytes. J. Cell Biol. 104:1389-1402.
Gumbiner, B., and K. Simons. 1986. A functional assay for proteins involved
in establishing an epithelial occluding barrier: identification ofa uvomorulinlike polypeptide. J. Cell Biol. 102:457--468.
Gumbiner, B., B. Stevenson, and A. Grimaldi. 1988. The role of the cell adhesion molecule uvomornlin in the formation and maintenance of the epithelial
junctional complex. 3". Cell Biol. 107:!575-1587.
Hamaguchi, M., N. Matsuyoshi, Y. Ohnishi, B. Gotoh, M. Takeichi, and Y.
Nagal. 1993. p60.... causes tyrosine phosphorylation and inactivation of
the N-cadherin-catenin cell adhesion system. EMBO (Eur. MoL Biol. Organ.) J. 12:307-314.
Herrenknecht, K., M. Ozawa, C. Eckerskorn, F. Lottspeich, M. Lenter, and
R. Kemler. 1991. The uvomornlin-anchorage protein a-catenin is a vinculin
homologue. Proc. Natl. Acad. Sci. USA. 88:9156-9160.
Hirano, S., N. Kimoto, Y. Shimoyama, S. Hirohashi, and M. Takeichi. 1992.
Identification of a neural ct-catenin as a key regulator of cadherin function
and multicellular organization. Cell. 70:293-301.
Hodivala, K. J., and F. M. Watt. 1994. Evidence that cadherins play a role in
the downregulation of integrin expression that occurs during keratinocyte
terminal differentiation. 3". Cell Biol. 124:589-600.
Itoh, M., S. Yonemura, A. Nagafuchi, Sa. Tsukita, and Sh. Tsukita. 1991. A
220-kD undercoat-constitutive protein: its specific localization at cadherinbased cell-cell adhesion sites. J. Cell Biol. 115:1449-1462.
Itoh, M., A, Nagafuchi, S. Yonemura, T. Kitani-Yasuda, Sa. Tsukita, and Sh.
Tsukita. 1993. The 220-kD protein colocalizing with cadherins in nonepithelial cells is identical to ZO-1, a tight junction-associated protein in epithelial ceils: cDNA cloning and immunoelectron miotoscopy. J. Cell Biol.
121:491-502.
Kemler, R. 1993. From cadherins to catenins: cytoplasmic protein interactions
and regulation of cell adhesion. Trends Genetics. 9:317-321.
Kinjo, M., K. Oka, S. Nalto, S. Kohga, K. Tanaka, S. Obushi, Y. Hayata, and
K. Yasumoto. 1979. Thromboplastic and fibrinolytic activities of cultured
human cancer cell lines. Br. J. Cancer. 39:15-22.
Knudsen, K. A., and M. J. Wheelock. 1992. Plakoglobin, or an 83-kD homologue distinct from #-catenin, interacts with E-cadherin and N-cadherin. J.
Cell Biol. 118:671-679.
Matsuyoshi, N., M. Hamaguchi, S. Taniguchi, A. Nagafuchi, S. Tsukita, and
M. Takeichi. 1992. Cadherin-mediated cell-cell adhesion is perturbed by
v-src tyrosine pbosphorylation in metastatic fibroblasts. J. Cell Biol. 118:
703-714.
Matsuzaki, F., R-M. M~ge, S. H. Jaffe, D. R. Friedlander, W. J. Gallin, J. I.
Goldberg, B. A. Cunningham, and G. M. Edelman. 1990. cDNAs of cell
adhesion molecules of different specificity induce changes in cell shape and
border formation in cultured S180 cells. J. Cell Biol. 110:1239-1252.
Mattey, D. L., and D. R. Garrod. 1986a. Calcium-induced desmosome formation in cultured kidney epithelial cells. J. Cell Sci. 85:95-111.
Mattey, D. L., and D. R. Garrod. 1986b. Splitting and internalization of the
desmosomes of cultured kidney epithelial cells by reduction in calcium concentration. J. Cell Sci. 85:113-124.
McCrea, P. D., and B. M. Gumbiner. 1991. Purification of a 92-kDa cytoplasmic protein tightly associated with the cell-cell adhesion molecule E-cadherin (Uvomorulin). J. Biol. Chem. 266:4514--4520.
McCrea, P. D., C. W. Turck, and B. Gumbiner. 1991. A homolog of the armadillo protein in Drosophila (plakoglobin) associated with E-cadherin.
Science (Wash. DC). 254:1359-1361.
McCrea, P. D., W. M. Brieher, and B. M. Gumbiner. 1993. Induction of a
secondary body axis in Xenopus by antibodies to/3-catenin. J. Cell Biol.
123:477--484.
McNeill, H., M. Ozawa, R. Kemler, and W. J. Nelson. 1990. Novel function
of the cell adhesion molecule uvomorulin as an inducer of cell surface polarity. Cell. 62:309-316.
Mege, R. M., F. Matsuzaki, W. J. Gallin, J. I. Goldberg, B. A. Cunningham,
and G. M. Edelman. 1988. Construction of epithelioid sheets by transfection
of mouse sarcoma cells with cDNAs for chicken cell adhesion molecules.
Proc. Natl. Acad. Sci. USA. 85:7274-7278.
Meyer, R. A., D. W. Laird, J.-P. Revel, and R. G. Johnson. 1992. Inhibition
of gap junction and adherens junction assembly by connexin and A-CAM antibodies. J. Cell Biol. 119:179-189.
Nagafuchi, A., and S. Tsukita. 1994. The loss of the expression of a catenin,
the 102kD cadherin-associated protein, in central nervous tissues during development. Dev. Growth Differ. 36:59-71.
Nagafuchi, A., M. Takeichi, and S. Tsukita. 1991. The 102kd cadherin-associated protein: similarity to vinculin and posttranscriptional regulation of expression. Cell. 65:849-857.
Watabe et al. Cadherin-Catenin in Epithelial Organization
255
We would like to thank Sh. Tsukita and M. Itoh for anti-ZO-1 antibody,
K . M . Yamada for anti-~l-integrin antibodies, D.R. Garrod for antidesmoplakin antibody, S. Shibamoto for anti-fl-catenin antibody, M.
Katayama for SHE78-7, and S. Hirano for c~N-catenin-transfected PC9. We
wish to thank T. Itoh for allowing us to use a confocal microscope and K.
Shimamura and T. Uemura for technical advice.
This work was supported by research grants from the Ministry of Education, Culture and Science of Japan, and the Human Frontier Science Program. M. Watabe is a recipient of a Fellowship of the Japan Society for the
Promotion of Science for Junior Scientists.
Received for publication 28 March 1994 and in revised form 24 June 1994.
Re'j~/'e/Ice$
Downloaded from on October 2, 2016
involved in control of tumorigenic growth through their association with APC. It should also be noted that/3-catenin is involved in axis determination in Xenopus embryos
(McCrea et al., 1993) and that its Drosophila homologue,
Armadillo, is essential for establishment of segment polarity
(Riggleman et al., 1990; Peifer and Wieschaus, 1990). Moreover, ~catenin is sensitive to extracellular and intracellular
signals; for example, tyrosine phosphorylation of this molecule is induced by treatment of cells with hepatocyte or epidermal growth factor (Shibamoto et al., 1994) as well as by
v-src transformation (Matsuyoshi et al., 1992; Hamaguchi et
al., 1993; Behrens et al., 1993). These findings suggest that
/3-catenin itself may play a signaling role in cell growth and
differentiation. Recent work suggests that cadherin-dependent cell contacts also regulate cell differentiation (Burdsal
et al., 1993) and integrin expression (Hodivala and Watt,
1994). These findings support the possibility that cadherins
and catenins contribute to signaling processes involved in
cell-cell interactions.
In conclusion, the cadherin-catenin system plays a central
role in structural and functional organization of epithelial
cell-cell contacts. The loss of cadherin activity impairs other
adhesion systems, and affects cell growth. Such effects may
be implicated in various abnormal behaviors of tumor cells,
in which cadherin activity is often down-regulated (Takeichi,
1993). Finally, it should be emphasized that the molecular
basis for the classical "contact inhibition" is not yet fully understood. Cadherin-mediated growth suppression is likely
related to this very important cellular behavior.
Published October 1, 1994
The Journal of Cell Biology, Volume 127, 1994
adhesiveness. Cancer Res. 52:5770-5774.
Shirayoshi, Y., A. Nose, K. Iwasaki, and M. Takeichi. 1986. N-linked
oligosaccharides are not involved in the function of a ceil-cell binding glycoprotein E-cadherin. Cell Structure & Function. 11:245-252.
Soriano, P., C. Montgomery, R. Geske, and A. Bradley. 1991. Targetted disruption of the c-src proto-oncogene leads to asteopetrosis in mice. Cell.
64:693-702.
Su, L-K., B. Vogelstein, and K. W. Kinzler. 1993. Association of the APC tumor suppressor protein with catenins. Science (Wash. DC). 262:1734-1737.
Takeichi, M. 1991. Cadherin cell adhesion receptors as a morphogenetic regulator. Science (Wash. DC). 251 : 1451- 1455.
Takeichi, M. 1993. Cadherins in cancer: implications for invasion and metastasis. Curr. Opin. Cell Biol. 5:806-811.
Tsukita, Sa., K. Oishi, T. Akiyama, Y. Yamanashi, T. Yamamoto, and Sh.
Tsukita. 1991. Specific proto-oneogeaic tyrosine kinases of src family are
enriched in cell-to-cell adherens junctions where the level of tyrosine phosphorylation is elevated. J. Cell Biol. 113:867-879.
Tsukita, Sh., M. Itoh, A. Nagafuchi, S. Yonemura, and Sa. Tsukita. 1993. Submembranous junctional plaque proteins include potential tumor suppressor
molecules. J. Cell Biol. 123:1049-1053.
Uchida, N., K. Shimamura, S. Miyatani, N. G. Copeland, D. J. Gilbert, N. A.
Jenkins, and M. Takeichi. 1994. Mouse aN-catenin: two isoforms, specific
expression in the nervous system, and chromosomal localization of the gene.
Dev. Biol. 163:75-85.
Volk, T., and B. G-eiger. 1986. A-CAM: a 135-kD receptor of intercellular
adberens junctions. II. Antibody-mediated modulation of junction formation. J. Cell Biol. 103:1451-1464.
Watabe, M., K. Matsumoto, T. Nakamura, and M. Takeichi. 1993. Effect of
hepatocyte growth factor on cadherin-mediated cell-cell adhesion. Cell
Structure & Function. 18:117-124.
Wheelock, M. J., and P. J. Jensen. 1992. Regulation of keratinocyte intercellular junction organization and epidermal morphogenesis by E-cadherin. J.
Cell Biol. 117:415-425.
Willott, E., M. S. Balda, A. S. Fanning, B. Jameson, C. Van Itallie, and J. M.
Anderson. 1993. The fight junction protein ZO-1 is homologous to the Drosophila discs-large tumor suppressor protein of septate junctions. Proc. Natl.
Acad. Sci. USA. 90:7834-7838.
Woods, D. F., and P. J. Bryant. 1991. The discs-large tumor suppressor gene
of Drosophila encodes a guanylate kinase homolog localized at septate junctions. Cell. 66:451-464.
256
Downloaded from on October 2, 2016
Navarro, P., M. Gomez, A. Pizarro, C. Gamallo, M. Quintaniila, and A. Cano.
1991. A Role for the E-cadherin cell-cell adhesion molecule during tumor
progression of mouse epidermal carcinogenesis. J. Cell Biol. 115:517-533.
Nose, A., A. Nagafuchi, and M. Takeichi. 1988. Expressed recombinant
cadherin mediate cell sorting in model systems. Cell. 54:993-1001.
Ozawa, M., M. Ringwald, and R. Kemler. 1990. Uvomorulin-catenincomplex
formation is regulated by a specific domain in the cytoplasmic region of the
cell adhesion molecule. Proc. Natl. Acad. Sci. USA. 87:4246-4250.
Parrish, E. P., P. V. Steart, D. R. Garrod, and O. R. Weller. 1987. Antidesmosomal monoclonal antibody in the diagnosis of intracranial tumors. J.
Pathol. 153:265-273.
Pasdar, M., and W. J. Nelson. 1988. Kinetics of desmosome assembly in
madin-darby canine kidney epithelial cells: temporal and spatial regulation
of desmoplakin organization and stabilization upon cell-cell contact. II. morphological analysis. J. Cell Biol. 106:687-695.
Peifer, M., and E. Wieschaus. 1990. The segment polarity gene armadillo encodes a functionally modular protein that is the Drosophila homolog of human plakoglobin. Cell. 63:1167-1178.
Piepenhagen, P. A., and W. J. Nelson. 1993. Defining E-cadherin-associated
protein complexes in epithelial cells: plakoglobin,/3- and ~-catenin are distinct components. J. Cell Sci. 104:751-762.
Riggleman, B., P. Schedl, and E. Wieschaus. 1990. Spatial expression of the
Drosophila segment polarity gene armadillo is posttranscriptionally regulated by wingless. Cell. 63:549-560.
Rodriguez-Boulan, E., and W. J. Nelson. 1989. Morphogenesis of the polarized epithelial cell phenotype. Science (Wash. DC). 245:718-725.
Rubinfeld, B., B. Souza, I. Albert, O. Miiller, S. H. Chamberlain, F. R.
Masiarz, S. Munemitsu, and P. Polakis. 1993. Association of the APC gene
product with/3-catenin. Science (Wash. DC). 262:1731-1734.
Shibamoto, S., M. Hayakawa, K. Takeuchi, T. Hori, N. Oku, K. Miyazawa,
N. Kitamura, M. Takeichi, and F. Ito. 1994. Tyrosine phosphorylation of
/~-catenin and plakogiobin enhanced by bepatocyte growth factor and epidermal growth factor in human carcinoma cells. Cell Adhesion & Comm.
1:295-305.
Shimoyama, Y., S. Hirohashi, S. Hirano, M. Noguchi, Y. Shimosato, M.
Takeichi, and O. Abe. 1989. Cadherin cell-adhesion molecules in human epithelial tissues and carcinomas. Cancer Res. 49:2128-2133.
Shimoyama, Y., A. Nagafuchi, S. Fujita, M. Gotoh, M. Takeichi, S. Tsukita,
and S, Hirohashi. 1992. Cadherin dysfunction in a human cancer cell line:
possible involvement of loss of c~-catenin expression in reduced cell-cell